Embed Size (px)
Citation preview
TitleThe predatory mite Neoseiulus womersleyi (Acari:Phytoseiidae) follows extracts of trails left by the two-spottedspider mite Tetranychus urticae (Acari: Tetranychidae).
Author(s) Shinmen, Tsubasa; Yano, Shuichi; Osakabe, Masahiro
Citation Experimental & applied acarology (2010), 52(2): 111-118
Issue Date 2010-10
URL http://hdl.handle.net/2433/129433
Right The original publication is available at www.springerlink.com
Type Journal Article
Textversion author
Kyoto University
brought to you by COREView metadata, citation and similar papers at core.ac.uk
provided by Kyoto University Research Information Repository
1
The predatory mite Neoseiulus womersleyi (Acari: Phytoseiidae) follows extracts 1
of trails left by the two-spotted spider mite Tetranychus urticae (Acari: 2
Tetranychidae) 3
4
Tsubasa Shinmen ▪ Shuichi Yano ▪ Masahiro Osakabe 5
6
T. Shinmen 7
Laboratory of Ecological Information, Faculty of Agriculture, Kyoto University, Sakyo-ku, Kyoto 8
606-8502, Japan 9
(Present address) Phamaceutical Research Laboratories No.2, Research & Development 10
Headquarters, Lion Corporation, Hirai 7-13-12, Edogawa-ku, Tokyo 132-0035; Japan 11
12
S. Yano* ▪ Mh. Osakabe 13
Laboratory of Ecological Information, Graduate School of Agriculture, Kyoto University, Sakyo-ku, 14
Kyoto 606-8502, Japan 15
*Corresponding Author 16
e-mail: [email protected] 17
Tel.: 81-75-753-6144 18
Fax: 81-75-753-614419
2
Abstract As it walks, the two-spotted spider mite Tetranychus urticae Koch (Acari: Tetranychidae) 20
spins a trail of silk threads, that is followed by the predatory mite, Neoseiulus womersleyi Schicha 21
(Acari: Phytoseiidae). Starved adult female N. womersleyi followed T. urticae trails laid down by 22
five T. urticae females but did not follow a trail of one T. urticae female, suggesting that the amount 23
of spun threads and their chemical components should correlate positively with the number of T. 24
urticae individuals. To examine whether chemical components of T. urticae trails are responsible for 25
the predatory mite’s trail following, we collected separate T. urticae threads from the exuviae and 26
eggs, and then washed the threads with methanol to separate chemical components from physical 27
attributes of the threads. Female N. womersleyi did not follow T. urticae trails that had been washed 28
with methanol but contained physical residues, but they did follow the direction to which the 29
methanol extracts of the T. urticae trails was applied. These results suggest that the predatory mite 30
follows chemical, not physical, attributes of T. urticae trails. 31
32
Key words Tetranychus urticae ▪ Neoseiulus womersleyi ▪ Silk thread ▪ Methanol extracts ▪ Trail 33
following 34
35
3
Introduction 36
Predatory mites are promising biological control agents against tetranychid mites (e.g. McMurtry 37
1992; Croft and Slone 1997). For effective use of predatory mites as biological control agents, the 38
prey-searching cues of the mites must be elucidated. Although volatiles produced by 39
spider-mite-infested plants are thought to attract predatory mites (Sabelis and Van de Baan 1983; 40
Dicke et al. 1990), recent observations in open environments suggest that spider mite patches on 41
plant leaves do not attract predatory mites at a distance on the plant (Zemek et al. 2008; Yano and 42
Osakabe 2009). On the other hand, Yano and Osakabe (2009) showed that the predatory mite 43
Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) follows trails left by adult female 44
Tetranychus urticae, suggesting that the spider mite trail could be a reliable prey-searching cue for 45
spider patches at a distance. 46
Tetranychus urticae feeds on various host plant species and is considered an important 47
4
agricultural pest globally (Jeppson et al. 1975). This mite often exhausts its host plant’s resources, 48
and mated adult females then disperse to new hosts, primarily by walking (Margolies and Kennedy 49
1985). Ambulatory dispersing adult female T. urticae follow the trails left by preceding females, 50
which results in aggregation at a new colony site (Yano 2008). Tetranychus urticae spins a trail of 51
silk threads when walking (Saito 1977). Silk threads spun by T. urticae have been reported to retain 52
predatory mites and elicit typical prey-searching behavior (Hislop et al. 1978; Hoy and Smilanick 53
1981; Hislop and Prokopy 1981). Hoy and Smilanick (1981) reported that the predatory mite 54
Metaseiulus (=Typhlodromus) occidentalis Nesbitt (Acari: Phytoseiidae) detects silk and other 55
residuals of T. urticae on spider-mite-infested leaves, which can be removed by washing the leaves 56
with water. Although they suggested that water-soluble extracts of deposits are likely to serve as a 57
search cue for the predatory mite, important cues may be physically washed off. Hislop and Prokopy 58
(1981) showed that although the predatory mite Neoseiulus fallasis (Garman) (Acari: Phytoseiidae) 59
5
preferred methanol extracts of T. urticae silk threads placed on filter paper, it showed a stronger 60
preference for intact silk. They concluded that N. fallasis may detect the chemical components of T. 61
urticae silk, although the physical stimulus of silk threads may also be involved. 62
However, these studies reported on the retention of predatory mites within a spider mite colony, 63
not on how they were guided toward the colony. Moreover, extracts of T. urticae silk contain 64
residuals, such as feces, exuviae, and even eggs, the latter of which are food of the above predatory 65
mites. Hence, these extracts may differ substantially from those of spider mite trails outside colonies. 66
To determine if predatory mites use the chemical components of spider mite trails to orient 67
themselves toward spider-mite-infested leaves, residual deposits that are apparently not associated 68
with trails outside the colony must be excluded. 69
In this study, we examined the prey-searching behavior of the endemic predatory mite 70
Neoseiulus womersleyi, which is an important predator of tetranychid mites in Japan (Hamamura 71
6
1986; Ehara and Shinkaji 1996). We collected T. urticae silk threads separately from their residuals. 72
Using extracts of the collected threads, we examined the hypothesis that N. womersleyi can detect 73
chemical components of T. urticae trails. 74
75
Materials and methods 76
Mites 77
The T. urticae study population was collected from a rose garden in Kyoto, Japan, and maintained on 78
individual discs of kidney bean Phaseolus vulgaris L. (Leguminosae) leaves pressed onto 79
water-saturated cotton in Petri dishes (90-mm diameter, 14 mm deep). The N. womersleyi study 80
population was also collected in Kyoto and was maintained on kidney bean leaf discs that were 81
heavily infested with T. urticae as prey. The leaf discs were maintained at 25 ± 2°C with 50 ± 5 % 82
relative humidity and a photoperiod of 16 : 8, light: dark (hereafter described as “laboratory 83
7
conditions”). 84
85
Preference of N. womersleyi for T. urticae trails 86
We first examined whether female N. womersleyi could follow trails laid down by T. urticae. The 87
term “trail”, as used in this study, refers to the silk threads and/or other chemical compound(s) 88
deposited by mated adult female T. urticae (hereafter described as “female T. urticae”). To conduct 89
dual-choice experiments under laboratory conditions, we connected one Parafilm square (Parafilm 90
M, American National Can Group, Chicago, IL, USA) and two bean leaf squares (10 10mm) with a 91
T-shaped Parafilm pathway on water-saturated cotton in Petri dishes (90-mm diameter, 14mm deep). 92
To induce a spider mite trail, we blocked a randomly selected branch with a piece of wet filte r paper, 93
and then introduced either one or five 2– to 4-day-old female T. urticae onto the Parafilm square 94
(Fig.1a). One to five adult female T. urticae correspond to a typical colony size of the mite in the 95
8
wild (Yano, unpublished results). After 1h, when all of the T. urticae females had moved to the 96
available bean leaf square, we removed the wet filter paper and two bean squares together with the T. 97
urticae females from each disc, leaving only a trail on the Parafilm (Fig.1b). 98
We used 1-day-old mated adult female N. womersleyi (hereafter described as “female N. 99
womersleyi”) that had been previously starved since late deutonymph period. To prepare these 100
females, we isolated an old deutonymph female and an adult male N. womersleyi in a 1.5-ml 101
microtube (Treff AG, Degersheim, Switzerland) with a water droplet. We had previously confirmed 102
that old deutonymph females mature without feeding and that these females most intensively follow 103
spider mite trails (Yano, unpublished results). To avoid cannibalism and unsuccessful mating, we 104
used mature females only when both the female and male were alive after 48h. We introduced 105
individual female mites to the bottom of the T-shaped pathway using a fine brush (Fig.1b) and 106
recorded which branch they followed to the far end. We used each female N. womersleyi and 107
9
T-shaped Parafilm path only once. The female N. womersleyi that did not reach either end within 108
5min were not included in the analysis. The number of replicates was 25 for trails made by five 109
females and 30 for trails made by one female. Experimental outcomes were compared using the 110
binomial tests (Sokal and Rohlf, 1995), with the common null hypothesis that a female N. 111
womersleyi would chose either of the two branches with equal probability (i.e. 0.5). 112
113
Preference of N. womersleyi for T. urticae trails washed with methanol 114
To conduct dual-choice experiments, trails made by five T. urticae females were induced on a 115
Parafilm pathway in the manner described above. We then transferred the Parafilm to a Petri dish 116
filled with methanol (Wako Pure Chemical Industries Ltd, Osaka, Japan, min. 99.5%) and gently 117
shook the dish for 5min. The Parafilm was then completely dried at room temperature and placed on 118
water-saturated cotton in another Petri dish (Fig.1b). To confirm whether threads were still present 119
10
on Parafilm pathways after they were washed with methanol, we observed threads on the surfaces of 120
randomly sampled Parafilm pathways using a scanning electron microscope (3D Real Surface View 121
Microscope VE-8800, Keyence, Osaka, Japan). 122
We then introduced a starved 1-day-old female N. womersleyi to the base point of the T-shaped 123
Parafilm in the manner described above (Fig.1b) and recorded the branch that the female first 124
followed to the far end. Each female N. womersleyi female and T-shaped Parafilm pathway was used 125
only once. Female N. womersleyi that did not reach an end point within 5min were not included in 126
the analysis. The number of replicates was 59. The numbers of females were compared using a 127
binomial test in the same manner described above. 128
129
Preference of N. womersleyi for T. urticae trail extracts 130
To collect T. urticae silk threads separately from residuals such as exuviae and eggs, we confined 10 131
11
T. urticae females within 6h of maturation in each of ten 1.5-ml microtubes (i.e. 100 females in total) 132
with a water droplet. We had previously confirmed that T. urticae females within 6h of maturation do 133
not oviposit when deprived of food. 134
To allow the females to deposit silk threads inside the microtubes, the tubes were kept under 135
laboratory conditions for 48h. Each tube was then opened onto a bean leaf disc to release the females. 136
After 30min, when all the females had exited, 80μl methanol was added to each tube and the tubes 137
were shaken gently for 20min using a constant temperature shaker (Synthetech Oven SO-1G, Nippon 138
Genetics Co., Tokyo, Japan) at 30°C. Extracts from the 10 tubes (c. 800μl total) were consolidated 139
into another tube and centrifuged at 12,000 r.p.m. for 5min. To exclude residual deposits, the top 140
clear layer in the tube was collected into another tube. To evaporate the methanol, the tube was 141
opened and placed in the shaker at 30°C for 60min. The final volume of the concentrated extracts was 142
c.400μl. For the control solvent, the same amount of methanol was poured into new microtubes, 143
12
which were treated in the manner described above. 144
To conduct the dual-choice experiments, we constructed T-shaped pathways of filter paper 145
(35 35mm, 2mm wide, Fig.2). Methanol extract of about 40 spider mite’s trails was applied to the 146
trunk and a randomly selected branch of each constructed pathway, with the control methanol applied 147
to the other branch. The two solutions were applied uniformly to both sides of the filter paper. Since 148
a starved N. womersleyi female is 0.25mm wide (1/8 of 2mm wide pathway) at most, N. womersleyi 149
advancing along the pathway would not contact more than 1/8 of the applied extracts. Since 1/8 of 40 150
T. urticae females is 5 females, this corresponds to the number of spider mites used in the first 151
experiment. Using 10μl micropipettes (Calibrated Pipets, Drummond Scientific Co., PA, USA), we 152
applied each solution little by little at the junction point to minimize mixing. The pathways were 153
completely dried at room temperature and then placed on three 15 x 15mm cardboard pieces fixed to 154
a Petri dish (Fig.2). We then introduced a starved 1-day-old female N. womersleyi to the base of the 155
13
T-shaped filter paper and recorded the branch that the females first followed to the far end. Females 156
that walked on the backside of the filter paper were observed via a 100 100-mm mirror placed 157
under the Petri dish (Fig.2). Each female N. womersleyi and T-shaped filter paper was used only once. 158
Female N. womersleyi that did not reach an end point within 5min were not included in the analysis. 159
The number of replicates was 54. The numbers of females were compared using a binomial test in the 160
same manner described above. 161
162
Results 163
Preference of N. womersleyi for T. urticae trails 164
Female N. womersleyi followed trails created by five female T. urticae (trail : control, 19 : 6; 165
P=0.0073, binomial test, Fig. 3a), but did not follow the trail made by one female T. urticae (trail : 166
control, 19 : 11; P = 0.100, binomial test, Fig. 3b). The latter statistically insignificant result, 167
14
however, may in part be due to the low number of replicates. 168
169
Preference of N. womersleyi for T. urticae trails washed with methanol 170
Although we found T. urticae threads on the Parafilm surfaces after they were washed with methanol, 171
the female N. womersleyi showed no preference for washed T. urticae trails (trail : control, 29 : 30; P 172
= 0.60, binomial test, Fig. 3c). 173
174
Preference of N. womersleyi for T. urticae trail extracts 175
Neoseiulus womersleyi females preferred trail branches with the methanol extract of T. urticae trails 176
to those with only methanol (trail extracts : methanol, 35 : 19; P = 0.020, binomial test, Fig. 3d), 177
suggesting that chemical components of T. urticae trails are used as a prey-searching cue. In addition, 178
we observed that some female N. womersleyi slowed down on the filter paper when they encountered 179
15
the trail extract. We also observed that some turned back from a trail-extract branch and retraced 180
their steps to the other, methanol-only branch. These behaviors were not observed during the 181
previous experiment that presented female N. womersleyi with T. urticae trails washed with 182
methanol. 183
184
Discussion 185
Not surprisingly, starved female N. womersleyi followed T. urticae trails in a density-dependent 186
fashion, as the amount of spun threads and their chemical components should correlate positively 187
with the number of T. urticae individuals used. Conspecific adult females (Yano 2008) and the 188
predatory mite Phytoseiulus persimilis (Yano and Osakabe 2009) also follow T. urticae trails in a 189
density-dependent fashion. 190
Because T. urticae spins silk threads while walking (Saito 1977), large quantities of threads can 191
16
be collected from long-settled mite colonies. However, residual deposits, such as exuviae, feces, 192
carcasses, and all stages of living mites, cannot be excluded from these samples. Moreover, as the 193
attractiveness of T. urticae threads appears to decline with time (Yano 2008), the properties of fresh 194
threads may differ from those collected at spider-mite colonies, which contain old threads. Therefore, 195
we examined T. urticae threads produced within 48 h, although some feces may have been included 196
in the extract. 197
The predatory mite did not follow T. urticae trails that had been washed with methanol, whereas 198
they did follow extracts of T. urticae trails. These results suggest that chemical, not physical 199
attributes of T. urticae trails are responsible for trail following by N. womersleyi, although the 200
washing treatment may have altered the physical properties of T. urticae threads. It is possible that 201
substances not soluble in methanol may be also responsible for the predatory mite’s trail following. 202
However, N. womersleyi females did not follow washed T. urticae trails where substances not soluble 203
17
in methanol should remain, suggesting that such substances are unimportant. Some N. womersleyi 204
females slowed down and exhibited typical prey-searching behavior (Hoy and Smilanick 1981) on 205
filter paper painted with extracts of T. urticae trails, but no predator engaged in this behavior on 206
trails washed with methanol. These observations further suggest that the predatory mite responds 207
positively to chemical components in the T. urticae trails as a prey-searching cue. Therefore, 208
predatory mites most likely use the chemical cues of T. urticae trails to find prey colonies, and then 209
use both physical and chemical trail cues after arriving at the colonies, as suggested by Hislop and 210
Prokopy (1981). 211
To efficiently attract and retain predatory mites of tetranychid mite colonies, the chemical 212
components responsible for trail following by predatory mites must be clarified. Methods to collect 213
efficiently the unadulterated spider-mite threads should also be improved. 214
215
Acknowledgements We thank K. Oku of the National Agricultural Research Center, R. Uesugi 216
18
of National Institute of Vegetable and Tea Science, and H. Iida of the National Agriculture and Food 217
Research Organization for helpful suggestions. We also thank D. E. Bowler and other members in our 218
laboratory for valuable suggestions. This study was partly supported by the Japan Society for the 219
Promotion of Science [Basic Research C; grant number 21580066 to S. Y.], by funding for the 220
development of new biorational techniques in sustainable agriculture, and by a Grant-in-Aid for the 21st 221
Century COE program for Innovative Food and Environmental Studies Pioneered by Entomomimetic 222
Sciences, from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. 223
224
References 225
Croft BA, Slone DH (1997) Equilibrium densities of European red mite (Acari: Tetranychidae) 226
after exposure to three levels of predacious mite diversity on apple. Environ Entomol 227
26:391-397 228
Dicke M, Sabelis MW, Takabayashi J, Bruin J, Posthumus MA. (1990) Plant strategies of 229
manipulating predator prey interactions through allelochemicals: prospects for application in 230
pest control. J Chem Ecol 16:3091-3118 231
Ehara S, Shinkaji N (1996) Principles of Plant Acarology. Zenkoku Noson Kyoiku Kyokai, Tokyo (in 232
Japanese). 233
Hamamura T (1986) Studies on the biological control of Kanzawa spider mite, Tetranychus kanzawai 234
Kishida by the chemical resistant predacious mite, Amblyseius longispinosus (Evans) in tea 235
fields (Acarina: Tetranychidae, Phytoseiidae). Bull Fruit Tree Res Stn E 21:121-201 (in 236
Japanese with English summary) 237
Hardman JM, Smith RF, Bent E (1995) Effects of different integrated pest management programs on 238
biological control of mites on apple by predatory mites (Acari) in Nova Scotia . Environ Entomol 239
24:125-142 240
19
Hislop RG, Alves N, Prokopy RJ (1978) Spider mite substances influencing searching behaviour of 241
the predator Amblyseius fallacies on apples. Fruit Notes 43:8-11 242
Hislop RG, Prokopy RJ (1981) Mite predator responses to prey and predator-emitted stimuli. J Chem 243
Ecol 7:895-904 244
Hoy MA, Smilanick JM (1981) Non-random prey location by the phytoseiid predator Metaseiulus 245
occidentails: differential responses to several spider mite species. Entomol Exp Appl 246
29:241-253 247
Jeppson LR, Keifer HH, Baker EW (1975) Mites Injurious to Economic Plants. University of 248
California Press, Berkeley, CA, USA. 249
Margolies DC, Kennedy GG (1985) Movement of the two-spotted spider mite Tetranychus urticae 250
Koch (Acari: Tetranychidae), among hosts in a corn-peanut agroecosystem. Entomol Exp Appl 251
37:55-61 252
McMurtry JA (1992) Dynamics and potential impact of generalist phytoseiids in agroecosystems and 253
possibilities for establishment of exotic species. Exp Appl Acarol 14:371-382 254
Sabelis MW, Van de Baan HE (1983) Location of distant spider mite colonies by phytoseiid 255
predators: demonstration of specific kairomones emitted by Tetranychus urticae and 256
Panonychus ulmi. Entomol Exp Appl 33:303-314 257
Saito Y (1977) Study on spinning behavior of spider mites (Acarina: Tetranychidae). I. Method for 258
quantitative evaluation of the mite webbing, and the relationship between webbing and walking. 259
Jpn J Appl Entomol Zool 21: 27-34 (in Japanese, with English summary) 260
Sokal RR, Rohlf FJ (1995) Biometry. The principles and practice of statistics in biological research, 261
3rd edn. Freeman, San Fransisco. 262
Yano S (2008) Collective and solitary behaviors of the two-spotted spider mite (Acari: 263
Tetranychidae) are induced by trail following. Ann Entomol Soc Am 101:247-252 264
20
Yano S, Osakabe Mh (2009) Do spider mite-infested plants and spider mite trails attract predatory 265
mites? Ecol Res 24:1173-1178 266
Zemek R, Nachman G, Ruzickova S (2008) How does Phytoseiulus persimilis find its prey when 267
foraging within a bean plant? In Bertrand M, Kreiter S, McCoy KD, Migeon A, Navajas M, 268
Tixier M-S, Vial L (eds) Integrative Acarology. Proceedings of the 6th European Congress. 269
European Association of Acarologists 2008, pp390-393 270
271
21
Figure Legends 272
Fig.1 Experimental designs for testing the preference of N. womersleyi females for T. urticae trails 273
and trails washed with methanol. a) To guide T. urticae females in one direction, a piece of wet filter 274
paper blocked the other path branch; b) A 1-day-old female N. womersleyi was placed on the base 275
point of the T-shaped Parafilm. We recorded the branch that the female first followed to the far end. 276
277
Fig.2 Experimental design for testing the preference of N. womersleyi for a methanol extract of T. 278
urticae trails. The methanol trail extract was applied to the trunk of a filter-paper path and a 279
randomly selected branch; pure methanol was applied as a control to the other branch. 280
281
Fig.3 Summary of the dual-choice experiments. Female N. womersleyi followed T. urticae trails laid 282
down by a) five females but did not follow a trail of b) one T. urticae female. Moreover, female N. 283
22
womersleyi c) did not follow T. urticae trails that had been washed with methanol but d) did follow 284
the methanol extracts of the T. urticae trails. 285
？
a) b)Wet filter paper(barrier)
10 x 10 mm bean leaf (trap)
Parafilm pathway(50 x 50 mm, 2mm wide)
Wet cotton(barrier)
T. urticaefemale
N. womersleyifemale
5 x 5 mmParafilm
Trail
Fig. 1
？
N. womersleyifemale
Petri dish (90-mm diameter)
Methanol solvent
Trail extractswith methanol
15 x 15mm card board
100 x 100 mm mirror
Fig. 2
Fig. 3
a) Trail (5 females)
b) Trail (1 female)
0 0.5 10.51
25
30
59
54
Fraction predators
Treatments Controls
c) Washed trails
d) Trail extracts
P=0.0073
NS (P=0.10)
NS (P=0.60)
P=0.020
